HUMAN MUTATION 29(4),537^544,2008
Branchio-Oto-Renal Syndrome (BOR): Novel
Mutations in the EYA1 Gene, and a Review
of the Mutational Genetics of BOR
Dana J. Orten,1Stephanie M. Fischer,2Jessica L. Sorensen,2Uppala Radhakrishna,1Cor W.R.J. Cremers,3
Henri A.M. Marres,3Guy Van Camp,4Katherine O. Welch,5Richard J.H. Smith,2and
William J. Kimberling1?
1Department of Genetics, Boys Town National Research Hospital, Omaha, Nebraska;2Department of Otolaryngology, Head and Neck Surgery,
University of Iowa, Iowa City, Iowa;3Department of Otorhinolaryngology, Radboud University Nijmegen Medical Centre, Nijmegen, The
Netherlands;4Department of Medical Genetics, University of Antwerp, Antwerp, Belgium;5Department of Biology, Gallaudet University,
Communicated by Maria Rita Passos-Bueno
Branchio-oto-renal syndrome (BOR) is an autosomal dominant disorder characterized by the association
of branchial and external ear malformations, hearing loss, and renal anomalies. The phenotype varies from ear
pits to profound hearing loss, branchial fistulae, and kidney agenesis. The most common gene mutated in BOR
families is EYA1, a transcriptional activator. Over 80 different disease-causing mutations have been published
(http://www.healthcare.uiowa.edu/labs/pendredandbor/, last accessed 20 November 2007). We analyzed the
EYA1 coding region (16 exons) from 435 families (345 at the University of Iowa [UI] and 95 at Boys Town
National Research Hospital [BTNRH], including five at both) and found 70 different EYA1 mutations in 89
families. Most of the mutations (56/70) were private. EYA1 mutations were found in 31% of families (76/248)
fitting established clinical criteria for BOR and 7% of families with questionable BOR phenotype (13/187).
Severity of the phenotype did not correlate with type of mutation nor with the domain involved. These results
add considerably to the spectrum of EYA1 mutations associated with BOR and indicate that the BOR
phenotype is an indication for molecular studies to diagnose EYA1-associated BOR. Hum Mutat 29(4),
rrrr2008 Wiley-Liss, Inc.
KEY WORDS: branchio-oto-renal syndrome; mutation analysis; genetic heterogeneity; EYA1.
Branchio-oto-renal syndrome (BOR) is a dominant disorder that
affects about 1 in 40,000 children, including 2% of profoundly deaf
children [Fraser et al., 1980]. The most common gene mutated in
BOR families is EYA1 [Abdelhak et al., 1997b], the human homolog
of the Drosophila eyes absent gene [Bonini et al., 1993]. Human
(EYA1; MIM] 601653) consists of 16 coding exons that extend over
156kb. At least three alternatively spliced transcripts differ only in
their 50regions [Abdelhak et al., 1997a; Kemperman et al., 2002b],
and an additional shorter transcript (variant 4: NM_172059.1)
excludes part of the coding region, skipping part of exon 7 and all of
exon 12 (see Supplementary Fig. S1, available online at http://
(transcript variant 3; NM_000503) encodes the longest transcript,
a 592–amino acid protein. Over 80 different disease-causing
mutations of EYA1 that result in either BOR or branchial-otic
syndrome (BO) have been published [Kumar et al., 1997, 1998;
Vincent et al., 1997; Abdelhak et al., 1997a, 1997b; Sanggaard et al.,
2007]. All known EYA1 mutations associated with BOR affect more
than one isoform. Mutations in exon 12 (skipped in the shortest
transcript variant 4) or its adjacent splice sites indicate that the
longer isoforms are necessary for EYA1 function. A list of BOR/BO
mutations is maintained at http://www.healthcare.uiowa.edu/labs/
Numerous polymorphisms of EYA1 have been reported
[Abdelhak et al., 1997a; Sanggaard et al., 2007]. When allelic
variants are discovered, it is not always clear whether they cause
disease. Since mutations in EYA1 are not found in 60% of people
with a BOR phenotype [Chang et al., 2004], caution must be used
when interpreting the effect of missense mutations in a single
family, especially if rigorous population-based studies have not
Published online 25 January 2008 in Wiley InterScience (www.
The Supplementary Material referred to in this article can be
accessed at http://www.interscience.wiley.com/jpages/1059-7794/
Received 31January 2007; acceptedrevisedmanuscript 31October
?Correspondence to:William J. Kimberling, Department of Genet-
ics, Boys Town National Research Hospital, 555 North 30th Street,
Omaha, NE 68164. E-mail: email@example.com
Grant sponsor: National Institutes of Health (NIH)^National Insti-
tute of Dental and Craniofacial Research (NIDCR); Grant number:
5R01DE014090-04; Grant sponsor: NIH^National Institute on Deaf-
ness and Other Communication Disorders (NIDCD); Grant numbers:
5R01DC003544-09; and 5R01DC04293.
rrrr2008 WILEY-LISS, INC.
The vertebrate eya gene family comprises four transcriptional
activators that ensure normal branchial arch and epibranchial
placode formation and sensory neurogenesis including hair cell and
neuron formation in the inner ear [Buller et al., 2001; Zou et al.,
2004; Brugmann and Moody, 2005; Fritzsch et al., 2006]. In the
kidney, Eya1 is a patterning gene essential for early metanephric
mesenchyme development [Dressler, 2006]. The structure of eya
proteins includes a conserved carboxy terminal eya-homologous
region (eyaHR) and a divergent proline-serine-threonine–rich
transactivation domain (eya variable region) [Zhang et al., 2004].
Studies in Drosophila indicate that the eyaHR mediates interac-
tions with so (sine oculis) and dac (dachshund) [Bui et al., 2000].
Expression of both eya and so is initiated by ey (eyeless) [Brodbeck
and Englert, 2004]. The vertebrate orthologs of so are the six gene
family, which bind to eya proteins, inducing nuclear translocation
of the complex [Ohto et al., 1999]. Functional analysis of human
EYA1 mutations in an in vivo Drosophila developmental system
suggests that defects in either phosphatase or transcriptional
activation occur [Mutsuddi et al., 2005]. Distinct defects in
protein function are predicted to lead to differences in phenotype.
Although the eyaHR domain was initially thought to be the site of
the majority of BOR mutations, mutations were later found
throughout the gene [Chang et al., 2004]. Three human
mutations in a second BOR gene (SIX1; MIM] 601205) associate
with BOR in four families [Ruf et al., 2004]. EYA1 and SIX1
proteins cooperate to activate the SALL1 promoter [Chai et al.,
2006]. Two families with mutations in (SALL1; MIM] 602218)
have a BOR-like phenotype lacking some of the characteristics of
Townes-Brocks syndrome [Engels et al., 2000; Albrecht et al.,
2004; Botzenhart et al., 2007].
The genotype–phenotype relationship for BOR and the Eya1
and Six1 genes is still poorly understood [Zhang et al., 2004].
Pathologic variation is associated with both missense and nonsense
mutations. The observation that some mutant proteins are rapidly
degraded suggests that haploinsufficiency causes BOR [Zhang
et al., 2004]. In large pedigrees, penetrance appears to be 100%,
although expressivity is highly variable [Chen et al., 1995; Chang
et al., 2004]. Recently, studies in Eya1 mutant mice found two loci
that suppress the BOR phenotype, suggesting the possible
existence of modifier genes in humans [Niu et al., 2006].
Genetic heterogeneity and the spectrum of phenotypes
associated with EYA1 mutations make the diagnosis of BOR
sometimes difficult. Thus, mutation screening of EYA1 is a
valuable clinical test. However, the sensitivity of mutation
screening to detect disease-causing allele variants of EYA1 is
technique-dependent. Direct sequencing or heteroduplex analysis
detects substitutions and small insertions or deletions, but will not
identify large deletions, duplications and chromosomal rearrange-
ments. These changes, which are believed to constitute about 20%
of EYA1 mutations in persons with BOR [Chang et al., 2004],
result in a ‘‘false negative’’ because only the normal allele is
amplified. In this study, 95 BOR families were screened for EYA1
mutations by direct sequencing and 345 families were screened by
heteroduplex analysis. In 56 families, exons with known
polymorphisms were sequenced in additional family members for
SNP analysis, to evaluate linkage or possible large deletions.
All procedures were approved by the Institutional Review Board
(IRB) at the University of Iowa (UI) or Boys Town National
Research Hospital (BTNRH). Phenotypes were evaluated by the
diagnostic criteria described previously [Chang et al., 2004]. Major
criteria are: branchial anomalies, deafness, preauricular pits, and
renal anomalies. Minor criteria are external ear anomalies, middle
ear anomalies, inner ear anomalies, preauricular tags, and others,
including facial asymmetry and palate abnormalities. To be
classified as BOR, an affected individual must have at least three
major criteria; two major criteria and at least two minor criteria; or
one major criterion and an affected first-degree relative meeting
the criteria for BOR. All families in this study had at least one of
the major criteria and most had two or more BOR-associated
abnormalities. Families are designated by letters indicating the
laboratory where the DNA was tested plus arbitrary numbers.
At BTNRH, genomic DNA was isolated from peripheral blood
samples using Puregene DNA Purification kits (Gentra, Minnea-
polis, MN). Most samples were PCR amplified and sequenced
(both directions) by Polymorphic DNA Technologies Inc.
(Alameda, CA; www.polymorphicdna.com/-reseqvardisc.html) on
an ABI (Applied Biosystems, Foster City, CA) 3730xl capillary
DNA Analyzer. Primers designed by Polymorphic are available
upon request. Sequences were analyzed using Mutation Surveyor
(SoftGenetics LLC, State College, PA). Mutation and SNP results
were organized and shown in pedigrees using Progeny Lab 6
(Progeny Software LLC, South Bend, IN). In some cases, mutant
alleles were cloned and sequenced after PCR amplification using a
TA cloning kit (Invitrogen, Carlsbad, CA). Mutations in BOR
families were confirmed by restriction endonuclease analysis
(Supplementary Table S1), or by sequencing with previously
published primers [Abdelhak et al., 1997a; Kumar et al., 1998].
Variants seen only in controls were detected in both forward and
reverse sequences. New primers (available upon request) were
designed for exon 11, to avoid SNPs within the primer, and exon
16, to avoid a poly A tract in IVS16.
At the Molecular Otolaryngology Research Laboratory, UI
(MORL), genomic DNA was extracted with FlexiGene DNA
purification kits (Qiagen, Valencia, CA). Amplified PCR products
were first screened by DHPLC heteroduplex analysis using the
Transgenomic WAVE system (Transgenomic, Omaha, NE) as
previously described [Prasad et al., 2004]. Samples with rare
DHPLC profiles were sequenced on an ABI model 3700
automated sequencer and analyzed using the Sequencer 4.1
software program package (Gene Codes, Ann Arbor, MI).
Previously published mutations were detected via heteroduplex
(BTNRH) or SSCP (MORL) on mutation detection enhancement
(MDEs) gels (Lonza, Walkersville, MD) as previously described in
the references footnoted in Table 1. Names of all variants were
checked using the Mutalyzer program (www.LOVD.nl/mutalyzer).
Our mutation numbering system corresponds to the longest
transcript variant: 3, NM_000503.3, and will differ from that
previous published. For that reason, we have included the
contextual surround of each pathologic variant (Supplementary
Table S1). Numbering according to this and other isoforms is
available at http://www.healthcare.uiowa.edu/labs/pendredandbor/.
Reference for intronic alterations is NC_000008.9. Nucleotide 11
is A of the ATG codon in the reference sequence according to
journal guidelines (www.hgvs.org/mutnomen).
At MORL, novel variants were tested by DHPLC in 57 control
samples. Samples with rare DHPLC profiles were sequenced as
described above. At BTNRH, all controls were evaluated by
538HUMAN MUTATION 29(4),537^544,2008
TABLE 1. EYA1Mutations in BOR Families?
Exon or IVS
Yes 0.42,3e ESE: (?)SF2/ASF;
c.639G4Cp.Q213H0.58,3e 08 UUU
2 yes;2 no
2 U,2 no
R-ESE: (1); ESE: (1)
p.R440Q0.50,5e 145F,1U,1S6 yes,1U2 U,2 yes,3 no
SC35; SS: (?)A
1yes 2 U
F, familial; S, sporadic; U, unknown; Invariant, mutation alters‘ag’or ‘gt’at the splice junction; ESE, mutation alters exonic splice enhancer binding sites for SF2/ASF, SC35,
SRp40, or SRp55; R-ESE (RESCUE-ESE), mutation altersenhancers found more frequently in exons with nonconsensus splice sites; SS (SpliceScan analysis), mutation alters
predicted score forsplice donor (D) orsplice acceptor (A); (1), improved score; (?), reduced score in ESE, R-ESE,or SpliceScan analysis (see Materials and Methods).
?The numbering system corresponds to the longest transcript: EYA1C (isoform 3, NM_000503.3). Nucleotide 11 is A of the ATG codon. Reference for intronic alterations is
NC_000008.9.Unlessnotedotherwise,themutationsarenoveltothisstudy. Mutations (DNAchangebold) andphenotypeinformation (bold) forfamiliesstudiedatBTNRHandMORL
were previously published by:cChanget al. ,dUsami et al. , andeKumaret al. [1997 ,1998]. Additional families with the same mutations were described by:fRickard et al.
,gFukuda et al. ,hSanggaard et al. ,iHenriksen et al. ,jNamba et al. , andkAbdelhak et al. [1997a,1997b].lPublished base change was corrected.
Additional detailed phenotype information, indicated by bold, was published by:mWiddershoven et al. ,nCremers and Fikkers-Van Noord ,oKemperman et al. [2002a],
pCremers et al. ,qCremers ,rCremers and Marres , andsKempermanet al. . Number of families includes this study and previously published families with the
samemutations. BOR phenotypiccriteriaare according toChang,et al.  (seeMaterials andMethods).
aAminoacid (AA) substitutionscores (0-1) arebasedonaminoacidproperties,withscoresbelow 0.5 indicatingasigni¢cantchange.Conseqscoresforaminoacidsarefrom1^9,
with 1 as not conserved, 4^5 as average, and 9 as highly conserved. Conseq residue functions are as follows: e, exposed; b, buried; f, functional (conserved, exposed); and s,
structural (conserved, buried).
bE¡ects on splicing wereevaluated for mutations that did not results in stop codons or frameshifts.
HUMAN MUTATION 29(4),537^544,2008 539
sequencing by Polymorphic DNA Technologies Inc. (see Mutation
Analysis, above). Controls included DNA from 63 to 74
unaffected married in spouses from BOR families (mostly from
the United States and Europe) plus 12 samples from an unrelated
study (for exons with common polymorphisms), or 84 control
samples (from the United States, mainly European ancestry) from
an unrelated study. All BTNRH controls were screened by
sequencing, which would detect novel mutations seen at MORL,
and we determined that novel BTNRH mutations were detectable
at MORL using DHPLC. Between the two groups, all rare variants
were tested in at least 132 control samples (264 alleles). To
compensate for possible interference from SNPs in DHPLC, exons
8, 11, 14, and 16, which contained novel mutations seen only by
BTNRH, were tested in both sets of BTNRH controls (156
samples, 312 alleles sequenced).
Our previously published mutations, including the common
missense mutation seen in seven families, c.1319G4A, p.R440Q,
had been tested by SSCP in 100 control samples at MORL [Chang
et al., 2004] and by heteroduplex analysis on MDEsgels in 100
control samples at BTHRH, for a total of 400 control alleles
negative for c.1319G4A, p.R440Q.
Prediction of Mutation Impact
Effects of mutations on predicted splicing were evaluated using
a neural network program from the Berkeley Drosophila Genome
Project (www.fruitfly.org/seq_tools/splice.html), and with the
recently developed SpliceScan program (http://bioinformatics.
ist.unomaha.edu/?achurban) [Churbanov et al., 2006]. Rare
missense or silent variants in exons were evaluated with ESEfinder
(http://rulai.cshl.edu/tools/ESE), which searches for exonic splice
enhancers (ESE) that are binding sites for specific serine/arginine-
rich splicing factors SF2/ASF, SC35, SRp40, and SRp55 [Cartegni
et al., 2003; Smith et al., 2006], and RESCUE-ESE (http:
//genes.mit.edu/burgelab/rescue-ese), which identifies enhancers
more frequent in exons with nonconsensus splice sites [Fairbrother
et al., 2004]. Missense mutations were evaluated for alterations in
secondary structure using the Lasergene suite of programs
(DNASTAR, Inc, Madison, WI) and for conservation using
ConSeq evaluation of the entire EYA1 amino acid sequence
(http://conseq.bioinfo.tau.ac.il) [Berezin et al., 2004]. Mutations
within conserved domains (CDs) were evaluated using a CD
multiple alignment link from a BLAST query (www.ncbi.nlm.-
nih.gov/BLAST). Normalized substitution scores for amino acid
changes (Table 1) are available (http://biochem218.stanford.edu/
Between the two groups, 70 different EYA1 mutations were
found in 89 families. A total of 54 are new mutations that have not
been published previously. Mutations were seen in all coding exons
except 03 (first coding exon) and 05 (Table 1). None of the
mutations were found in control samples. Results of mutation
testing at BRNRH and MORL are summarized in Table 2. Novel
mutations found at BTNRH in samples that had not been analyzed
by DHPLC were detectable by DHPLC at MORL, but SNPs in
some amplicons (Supplementary Table S2) complicated the
Most of the mutations (56/70) are
private. A total of nine mutations were seen in multiple families in
this study. Five more mutations had been found in additional
published families [Abdelhak et al., 1997a, 1997b; Rickard et al.,
2000; Fukuda et al., 2001; Namba et al., 2001; Henriksen et al.,
2004; Sanggaard et al., 2007]. The three most common mutations
in EYA1 are c.1319G4A: p.R440Q (seven families), c.889C4T,
p.R297X (six families, three in this study), and c.922C4T,
p.R308X (four families). These mutations were seen in multiple
ethnic groups, and two (c.889C4Tand c.1319G4A were seen in
both sporadic and familial cases, suggesting mutational hot spots.
All three base changes are C4T transitions (or G4A in the
reverse strand) at CpG dinucleotides. This type of mutation is
5–10 times as common in vertebrate genomes, due to 5 methylC
deamination, effects of flanking regions, and local GC content
[Zhao and Boerwinkle, 2002; Zhao and Jiang, 2007]. All three
sites had high scores in the MutPred program (Supplementary
Fig. S2), which takes into account flanking sequence and changes
in amino acid properties [Cooper and Krawczak, 1990; Krawczak
et al., 1998]. One less common CpG mutation, c.164C4T (one
family), had a low MutPred score, and two mutations had a similar
high score, c.880C4T (one family) and c.1081C4T (two
families), suggesting that additional sequence features may affect
mutation recurrence. Our three families with the c.889C4T
mutation are sporadic, with at least two different SNP profiles
(Table 3). For the most common mutation, c.1319G4A, we
observed at least three EYA1 SNP profiles (Table 3) for the seven
families, supporting our hypothesis of multiple origins for the base
Predicted e¡ects on splicing.
stop mutations would result in nonsense mediated decay, or
produce unstable or truncated proteins. A total of 12 splice
mutations are in the invariant ‘‘ag’’ or ‘‘gt,’’ predicted to cause
exon skipping. Two splice mutations create a new splice site with a
higher SpliceScan score (Table 1). Of the 13 missense variants, 11
alter predicted splicing and/or splice enhancers. c.638A4T and
c.639G4C reduce SpliceScan scores for both donor and acceptor
and c.1475G4C and c.1538T4C reduce the SpliceScan score for
the normal acceptor site. Splice enhancer effects include
disruption of an ESE for SF2/ASF by c.638A4T, disruption of
SF2/ASF and SC35 sites by c.1475G4C, and creation of a SRp55
ESE motif by c.639G4C, which may cause competition by SRp55
with splice factors that normally bind. Although the SpliceScan
sites do not change, 7 of the 13 missense mutations alter either
ESEs or RESCUE-ESEs. Experimental verification would be
needed to determine whether the missense or splice changes are
more important for altering the protein.
Predicted impact of missense mutations.
region, 8 of the 13 missense mutations (p.D429G, p.R440Q,
p.R492P p.V512F, p.L513P , p.Y527N, p.Y527C, and p.V550E) and
the in-frame deletion (p.553delV) are conserved in EYA1
homologs, and p.S487P is partially conserved (Fig. 1). Four
The 22 frameshifts and 20
In the eyaHR
TABLE 2. Summary of EYA1MutationTesting in BOR Families
Families with EYA1
aBORcriteria are according toChanget al. .
bIndividual laboratory totals and percentages include ¢ve families tested at both la-
boratories, two of which had their mutations found at both BTNRH and MORL.
540HUMAN MUTATION 29(4),537^544,2008
additional missense mutations (p.T55M, p.Q213L, p.I326N, and
p.E380L) are significant changes in amino acid properties,
indicated by substitution scores of r0.5, and are predicted to
change protein folding. Although p.R440Q is borderline for
Conseq and substitution scores, its conservation among EYA1
homologs (Fig. 1), and its association with BOR in seven families
indicates that it is pathologic. The Conseq and substitution scores
for p.Q213H do not indicate a significant change in protein
function, suggesting that altered splicing due to the c.639G4C
base change (see above) is more important than the missense
change. Eya1 catalytic domain phosphatase activity was severely
reduced with mouse p.S486P , homologous to human p.S487P
[Rayapureddi and Hegde, 2006] providing additional evidence
that the proline substitution is pathologic. Moreover, the ConSeq
score and alignment with eya1 proteins from other animals (Fig. 1)
indicates that some variation is tolerated at amino acid 487. A
nonpathologic variant, p.S487L, was seen in both a BOR family
and a control sample (Supplementary Table S3).
Similar to previous reports [Cremers and Fikkers-Van Noord,
1980], the syndrome was usually familial (60/89 families).
Phenotypes of individual families are summarized in Supplemen-
tary Table S1. The severity of phenotype does not correlate with
the type of mutation and is extremely variable even within
families. Although 44 out of 89 families in our study with EYA1
mutations had members with known renal symptoms or abnorm-
alities (Table 1; Supplementary Table S1), normal kidneys were
often seen in other affected members of the same families, and in
many cases, kidneys had not been clinically evaluated. In some
cases, kidney malformations were detected by urography, but renal
function was normal [Widdershoven et al., 1983]. Detailed clinical
evaluation has been published (see footnotes to Table 1) for five
BOR families that now have EYA1 mutations found [Cremers and
Marres, 1977; Cremers and Fikkers-Van Noord, 1980; Widder-
shoven et al., 1983; Cremers, 1983; Cremers et al., 1993; Usami
et al., 1999; Kemperman et al., 2001, 2002a].
SNPs and RareVariants with Unknown Function
Effects of hypomorphic variants in the ‘‘normal’’ allele
are unknown. Supplementary Table S2 shows SNPs (frequency
of Z2% in controls). SNPs in IVS07 and IVS11 are located in
published primers [Abdelhak et al., 1997a], and would interfere
with amplification of both alleles in some families. We are in the
process of evaluating the 10 variants with a frequency 410% for
correlation with severity of the phenotype in large families.
Supplementary Table S3 shows rare variants that are not
associated with BOR in the families, or that do not significantly
change the predicted protein structure. One variant, c.1359A4G,
could not be evaluated by SpliceScan or the neural network
program for its effects on splicing. It is the last base of exon 14,
next to a noncanonical ‘‘gc’’ splice site [Abdelhak et al., 1997a],
which is not predicted by current splice analysis programs. The
variant disrupts a RESCUE-ESE, a splice enhancer associated with
weak splice sites [Fairbrother et al., 2004]. Modification of such
splice enhancer sites may result in either exon skipping or the
addition of unwanted sequence to the mRNA. Since there is
strong evidence that altered EYA1 produces BOR under
conditions of haploinsufficiency, a sufficient reduction in the
amount of valid EYA1 transcript would be predicted to cause
BOR. In mice, for example, an intracisternal A particle insertion
mutation results in a recessive BOR phenotype [Johnson et al.,
1999], by causing improper splicing of 50% of the EYA1 mRNA.
The BOR family with the c.1359A4G variant has a dominant
inheritance pattern; the variant is heterozygous in the affected
parent and child, and not found in the unaffected parent or
siblings. Further studies of mRNA levels in lymphoblastoid cells
from family members (such as carried out by Vervoort et al.
) would be required to prove this prediction.
TABLE 3. SNPS AssociatedWith Common Mutations?
AmpliconSNPRefSeq#BT1513BT1469 BT1394MORL18 MORL31MORL40 BT1526 BT4038MORL15
?Recurrent mutations, numbered according to isoform 3, NM_000503.3 are indicated above the family numbers. All SNPs had a rare allele frequency Z2% in BTNRH controls
(see SupplementaryTable S3). Single bases indicate homozygosity, or association of the SNP with the mutant allele in multiple a¡ected family members. Underlined values
indicate the ¢rst SNPalleleassociatedwith the pathologic mutations; boldvalues indicate a second SNPallele associatedwith the pathologic mutation.
FIGURE 1. Conservation of amino acids altered by missense mu-
of human EYA1 sequence in human EYA4 and in Eya1 proteins
from rodents, zebra¢sh,chicken, frog, and dog is shown. Amino
acids that match humanEYA1are shaded, and mutations are in-
dicated above the aligned sequence, with the amino acid boxed
in the alignment. Except for p.S487, which is conserved in
EYA4, rodents, and zebra¢sh, all missense changes occurred at
amino acidsthat areconservedinalleightsequences. [Color ¢g-
HUMAN MUTATION 29(4),537^544,2008541
A rare missense variant (c.585A4G, p.I195M) was seen in one
proband with unknown family history. Although the variant does
not have a significant Conseq or substitution score, it does create a
new ESE for splicing factor SC35 and could be considered to be
Pathology of the other rare variants is extremely unlikely
because they either have no predicted effect, they were seen in
controls or unaffected family members, and/or there was another
more obvious pathologic mutation observed in the family.
The next step will be to test for large deletions, insertions or
complex rearrangements in EYA1 by multiplex ligation dependent
probe amplification (MLPA). SNP analysis has indicated at least
five BTNRH families likely to have EYA1 mutations that cannot
be detected by conventional sequencing, including the large
Dutch family used to refine the EYA1 locus [Cremers and Fikkers-
Van Noord, 1980; Kumar et al., 1992]. Three MORL BOR
families had previously published large deletions or complex
rearrangements [Chang et al., 2004] and testing of the remainder
of the families is in progress.
Testing for mutations in SIX1, 4, and 6, which occur as a gene
cluster [Ruf et al., 2004], is in progress and 10 SIX1 mutations
have been found in our families without detectable EYA1
mutations (Kochhar et al., in press). Additional candidate genes
are being tested via collaboration. The remaining families with
several affected individuals and no mutations detected will be a
valuable resource for identification of novel locus/loci responsible
for BOR, providing new insights into developmental pathways for
branchial arches, inner ear and kidney.
Effects of the 11 EYA1 SNPs with rare allele frequencies 410%
have not yet been fully evaluated. Combined with a null allele in
EYA1 or a gene in the same pathway, minor effects of SNPs on
protein function could increase BOR severity. A mouse eya1 allele,
which is improperly spliced 50% of the time [Johnson et al., 1999]
causes a ‘‘recessive’’ BOR that is affected by modifier genes [Niu
et al., 2006]. Similar variants may occur in humans. As a future
project, the combined family resources at BTNRH and MORL will
be used in an attempt to identify SNPs in the BOR genes that are
insufficient to cause BOR, but modify the severity of the disorder.
For nine of the SNPs, data in GenBank indicate that the common
variant differs depending on population (Supplementary Table S2).
The EYA1-SIX gene pathway is implicated in some types of cancer
[Li et al., 2002], and SNPs could also affect susceptibility to
tumors or other disorders.
Environmental factors, particularly the nutritional status of the
mother, have been shown to profoundly affect developmental
disorders. More than 10 years ago, folic acid supplements were
shown to prevent neural tube disorders such as spina bifida [Botto
et al., 1999]. Recent population studies show an association of
folic acid supplementation with reduced liability to cleft lip
[Wilcox et al., 2007] as well as other disorders including kidney
agenesis [Canfield et al., 2005]. The hearing of children with BOR
who have malformations of the inner ear, particularly enlarged
vestibular aqueduct (EVA), can be exacerbated by minor
playground injuries. Childhood infections increase the impact of
fistulas and inner ear malformations. People with BOR may be
born with minor kidney malformations, but suffer kidney
malfunction or failure as adults. Nutrition, drug treatment for
other diseases, or pregnancies might be predicted to contribute to
kidney failure. Family history and lifestyle evaluation in people
with BOR could lead to nutritional and lifestyle strategies that
prevent severe effects of BOR. For example, diagnosis of a BOR
gene mutation as the cause of hearing loss should lead to testing
for EVA and other ear malformations in which the hearing loss can
be exacerbated by the child’s environment (i.e., head trauma); it
might also be sensible to monitor kidney function and avoid
We are grateful to the families for their cooperation and
participation in the study. We thank Virginia W. Norris, MGC, and
Kathleen S. Arnos, PhD, at Gallaudet University, and Arti Pandya,
MD, at Virginia Commonwealth University, for recruitment and
evaluation of families. Supported by National Institutes of
Health (NIH)–National Institute of Dental and Craniofacial
Research (NIDCR) grant 5R01DE014090-04 (to W.J.K.) and
NIH–National Institute on Deafness and Other Communication
Disorders (NIDCD) grants 5R01DC003544-09 (to R.J.H.S.) and
5R01DC04293 (to K.O.W.). We thank Alexander Churbanov for
helpful discussions of splice site analysis, and Carol Carney, Maren
Jensen, and Kristal Platt, at BTNRH, and Amit Kochhar, at UI, for
Abdelhak S, Kalatzis V, Heilig R, Compain S, Samson D, Vincent C, Levi-
Acobas F, Cruaud C, Le MM, Mathieu M, Konig R, Vigneron J,
Weissenbach J, Petit C, Weil D. 1997a. Clustering of mutations
responsible for branchio-oto-renal (BOR) syndrome in the eyes absent
homologous region (eyaHR) of EYA1. Hum Mol Genet 6:2247–2255.
Abdelhak S, Kalatzis V, Heilig R, Compain S, Samson D, Vincent C, Weil
D, Cruaud C, Sahly I, Leibovici M, Bitner-Glindzicz M, Francis M,
Lacombe D, Vigneron J, Charachon R, Boven K, Bedbeder P , Van RN,
Weissenbach J, Petit C. 1997b. A human homologue of the Drosophila
eyes absent gene underlies branchio-oto-renal (BOR) syndrome and
identifies a novel gene family. Nat Genet 15:157–164.
Albrecht B, Liebers M, Kohlhase J. 2004. Atypical phenotype and
intrafamilial variability associated with a novel SALL1 mutation. Am J
Med Genet A 125:102–104.
Berezin C, Glaser F, Rosenberg J, Paz I, Pupko T, Fariselli P , Casadio R, Ben-
Tal N. 2004. ConSeq: the identification of functionally and structurally
important residues in protein sequences. Bioinformatics 20:1322–1324.
Bonini NM, Leiserson WM, Benzer S. 1993. The eyes absent gene: genetic
control of cell survival and differentiation in the developing Drosophila
eye. Cell 72:379–395.
Botto LD, Moore CA, Khoury MJ, Erickson JD. 1999. Neural-tube defects.
N Engl J Med 341:1509–1519.
Botzenhart EM, Bartalini G, Blair E, Brady AF, Elmslie F, Chong KL,
Christy K, Torres-Martinez W, Danesino C, Deardorff MA, Fryns JP ,
Marlin S, Garcia-Minaur S, Hellenbroich Y, Hay BN, Penttinen M,
Shashi V, Terhal P , Van ML, Whiteford ML, Zackai E, Kohlhase J. 2007.
Townes-Brocks syndrome: twenty novel SALL1 mutations in sporadic
and familial cases and refinement of the SALL1 hot spot region. Hum
Brodbeck S, Englert C. 2004. Genetic determination of nephrogenesis: the
Pax/Eya/Six gene network. Pediatr Nephrol 19:249–255.
Brugmann SA, Moody SA. 2005. Induction and specification of the
vertebrate ectodermal placodes: precursors of the cranial sensory organs.
Biol Cell 97:303–319.
Bui QT, Zimmerman JE, Liu H, Bonini NM. 2000. Molecular analysis of
Drosophila eyes absent mutants reveals features of the conserved Eya
domain. Genetics 155:709–720.
Buller C, Xu X, Marquis V, Schwanke R, Xu PX. 2001. Molecular effects of
Eya1 domain mutations causing organ defects in BOR syndrome. Hum
Mol Genet 10:2775–2781.
Canfield MA, Collins JS, Botto LD, Williams LJ, Mai CT, Kirby RS, Pearson
K, Devine O, Mulinare J. 2005. Changes in the birth prevalence of
542HUMAN MUTATION 29(4),537^544,2008
selected birth defects after grain fortification with folic acid in the
United States: findings from a multi-state population-based study. Birth
Defects Res A Clin Mol Teratol 73:679–689.
Cartegni L, Wang J, Zhu Z, Zhang MQ, Krainer AR. 2003. ESEfinder:
a web resource to identify exonic splicing enhancers. Nucleic Acids Res
Chai L, Yang J, Di C, Cui W, Kawakami K, Lai R, Ma Y. 2006.
Transcriptional activation of the SALL1 by the human SIX1 home-
odomain during kidney development. J Biol Chem 281:18918–18926.
Chang EH, Menezes M, Meyer NC, Cucci RA, Vervoort VS,
Schwartz CE, Smith RJ. 2004. Branchio-oto-renal syndrome: the
mutation spectrum in EYA1 and its phenotypic consequences. Hum
Chen A, Francis M, Ni L, Cremers CW, Kimberling WJ, Sato Y, Phelps PD,
Bellman SC, Wagner MJ, Pembrey M. 1995. Phenotypic manifestations
of branchio-oto-renal syndrome. Am J Med Genet 58:365–370.
Churbanov AE, Rogozin IB, Deogun JS, Ali HH. 2006. Method of
predicting splice sites based on signal interactions. Biol Direct 1:10.
Cooper DN, Krawczak M. 1990. The mutational spectrum of single base-
pair substitutions causing human genetic disease: patterns and
predictions. Hum Genet 85:55–74.
Cremers CW, Marres EH. 1977. [Deafness and hypacusis combined with
branchiogenic and sometimes renal congenital abnormalities]. Ned
Tijdschr Geneeskd 121:1676–1679.
Cremers CW, Fikkers-Van Noord M. 1980. The earpits-deafness syndrome.
Clinical and genetic aspects. Int J Pediatr Otorhinolaryngol 2:309–322.
Cremers CW. 1983. Congenital pre-auricular fistula communicating with
the tympanic cavity. J Laryngol Otol 97:749–753.
Cremers CW, Marres HA, Brunner HG. 1993. Neo-oval window technique
and myringo-chorda-vestibulopexy in the BOR syndrome. Laryngoscope
Dressler GR. 2006. The cellular basis of kidney development. Annu Rev
Cell Dev Biol 22:509–529.
Engels S, Kohlhase J, McGaughran J. 2000. A SALL1 mutation causes
a branchio-oto-renal syndrome-like phenotype. J Med Genet 37:
Fairbrother WG, Yeo GW, Yeh R, Goldstein P , Mawson M, Sharp PA, Burge
CB. 2004. RESCUE-ESE identifies candidate exonic splicing enhancers
in vertebrate exons. Nucleic Acids Res 32:W187–W190.
Fraser FC, Sproule JR, Halal F. 1980. Frequency of the branchio-oto-renal
(BOR) syndrome in children with profound hearing loss. Am J Med
Fritzsch B, Beisel KW, Hansen LA. 2006. The molecular basis of
neurosensory cell formation in ear development: a blueprint for hair
cell and sensory neuron regeneration? Bioessays 28:1181–1193.
Fukuda S, Kuroda T, Chida E, Shimizu R, Usami S, Koda E, Abe S,
Namba A, Kitamura K, Inuyama Y. 2001. A family affected by branchio-
oto syndrome with EYA1 mutations. Auris Nasus Larynx 28(Suppl):
Henriksen AM, Tumer Z, Tommerup N, Tranebjaerg L, Larsen LA. 2004.
Identification of a novel EYA1 splice-site mutation in a Danish
branchio-oto-renal syndrome family. Genet Test 8:404–406.
Johnson KR, Cook SA, Erway LC, Matthews AN, Sanford LP , Paradies NE,
Friedman RA. 1999. Inner ear and kidney anomalies caused by IAP
insertion in an intron of the Eya1 gene in a mouse model of BOR
syndrome. Hum Mol Genet 8:645–653.
Kemperman MH, Stinckens C, Kumar S, Huygen PL, Joosten FB, Cremers
CW. 2001. Progressive fluctuant hearing loss, enlarged vestibular
aqueduct, and cochlear hypoplasia in branchio-oto-renal syndrome.
Otol Neurotol 22:637–643.
Kemperman MH, Koch SM, Joosten FB, Kumar S, Huygen PL, Cremers
CW. 2002a. Inner ear anomalies are frequent but nonobligatory features
of the branchio-oto-renal syndrome. Arch Otolaryngol Head Neck Surg
Kemperman MH, Stinckens C, Kumar S, Joosten FB, Huygen PL, Cremers
CW. 2002b. The branchio-oto-renal syndrome. Adv Otorhinolaryngol
Kochhar A, Orten DJ, Sorensen JL, Fischer SM, Cremers CWRJ,
Kimberling WJ, Smith RJH. SIX1 mutation screening in 247
Branchio-oto-renal syndrome families: a recurrent missense mutation
associated with BOR. Hum Mutat (in press).
Krawczak M, Ball EV, Cooper DN. 1998. Neighboring-nucleotide effects on
the rates of germ-line single-base-pair substitution in human genes. Am J
Hum Genet 63:474–488.
Kumar S, Kimberling WJ, Kenyon JB, Smith RJ, Marres HA, Cremers CW.
1992. Autosomal dominant branchio-oto-renal syndrome–localization of
a disease gene to chromosome 8q by linkage in a Dutch family. Hum Mol
Kumar S, Deffenbacher K, Cremers CW, Van Camp G, Kimberling WJ.
1997. Branchio-oto-renal syndrome: identification of novel mutations,
molecular characterization, mutation distribution, and prospects for
genetic testing. Genet Test 1:243–251.
Kumar S, Kimberling WJ, Weston MD, Schaefer BG, Berg MA, Marres HA,
Cremers CW. 1998. Identification of three novel mutations in human
EYA1 protein associated with branchio-oto-renal syndrome. Hum Mutat
Li CM, Guo M, Borczuk A, Powell CA, Wei M, Thaker HM, Friedman R,
Klein U, Tycko B. 2002. Gene expression in Wilms’ tumor mimics the
earliest committed stage in the metanephric mesenchymal-epithelial
transition. Am J Pathol 160:2181–2190.
Mutsuddi M, Chaffee B, Cassidy J, Silver SJ, Tootle TL, Rebay I. 2005.
Using Drosophila to decipher how mutations associated with human
branchio-oto-renal syndrome and optical defects compromise the
protein tyrosine phosphatase and transcriptional functions of eyes
absent. Genetics 170:687–695.
Namba A, Abe S, Shinkawa H, Kimberling WJ, Usami SI. 2001. Genetic
features of hearing loss associated with ear anomalies: PDS and EYA1
mutation analysis. J Hum Genet 46:518–521.
Niu H, Makmura L, Shen T, Sheth SS, Blair K, Friedman RA.
2006. Identification of two major loci that suppress hearing loss
and cochlear dysmorphogenesis in Eya1bor/bor mice. Genomics 88:
Ohto H, Kamada S, Tago K, Tominaga SI, Ozaki H, Sato S, Kawakami K.
1999. Cooperation of six and eya in activation of their target genes
through nuclear translocation of Eya. Mol Cell Biol 19:6815–6824.
Prasad S, Kolln KA, Cucci RA, Trembath RC, Camp GV, Smith RJ. 2004.
Pendred syndrome and DFNB4-mutation screening of SLC26A4 by
denaturing high-performance liquid chromatography and the identifica-
tion of eleven novel mutations. Am J Med Genet 124A:1–9.
Rayapureddi JP , Hegde RS. 2006. Branchio-oto-renal syndrome associated
mutations in Eyes Absent 1 result in loss of phosphatase activity. FEBS
Rickard S, Boxer M, Trompeter R, Bitner-Glindzicz M. 2000. Importance of
clinical evaluation and molecular testing in the branchio-oto-renal
(BOR) syndrome and overlapping phenotypes. J Med Genet 37:
Ruf RG, Xu PX, Silvius D, Otto EA, Beekmann F, Muerb UT, Kumar S,
Neuhaus TJ, Kemper MJ, Raymond RM, Jr, Brophy PD, Berkman J,
Gattas M, Hyland V, Ruf EM, Schwartz C, Chang EH, Smith RJ,
Stratakis CA, Weil D, Petit C, Hildebrandt F. 2004. SIX1 mutations
cause branchio-oto-renal syndrome by disruption of EYA1-SIX1-DNA
complexes. Proc Natl Acad Sci USA 101:8090–8095.
Sanggaard KM, Rendtorff ND, Kjaer KW, Eiberg H, Johnsen T, Gimsing S,
Dyrmose J, Nielsen KO, Lage K, Tranebjaerg L. 2007. Branchio-oto-
renal syndrome: detection of EYA1 and SIX1 mutations in five out of six
Danish families by combining linkage, MLPA and sequencing analyses.
Eur J Hum Genet 15:1121–1131.
Smith PJ, Zhang C, Wang J, Chew SL, Zhang MQ, Krainer AR. 2006.
An increased specificity score matrix for the prediction of SF2/
ASF-specific exonicsplicing enhancers.
Usami S, Abe S, Shinkawa H, Deffenbacher K, Kumar S, Kimberling WJ.
1999. EYA1 nonsense mutation in a Japanese branchio-oto-renal
syndrome family. J Hum Genet 44:261–265.
Vervoort VS, Smith RJ, O’Brien J, Schroer R, Abbott A, Stevenson RE,
Schwartz CE. 2002. Genomic rearrangements of EYA1 account for a
large fraction of families with BOR syndrome. Eur J Hum Genet
HUMAN MUTATION 29(4),537^544,2008 543
Vincent C, Kalatzis V, Abdelhak S, Chaib H, Compain S, Helias J,
Vaneecloo FM, Petit C. 1997. BOR and BO syndromes are allelic defects
of EYA1. Eur J Hum Genet 5:242–246.
Widdershoven J, Monnens L, Assmann K, Cremers C. 1983. Renal
disorders in the branchio-oto-renal syndrome. Helv Paediatr Acta
Wilcox AJ, Lie RT, Solvoll K, Taylor J, McConnaughey DR, Abyholm F,
Vindenes H, Vollset SE, Drevon CA. 2007. Folic acid supplements and
risk of facial clefts: national population based case-control study. BMJ
Zhang Y, Knosp BM, Maconochie M, Friedman RA, Smith RJ. 2004. A
comparative study of Eya1 and Eya4 protein function and its implication
in branchio-oto-renal syndrome and DFNA10. J Assoc Res Otolaryngol
Zhao Z,Boerwinkle E.2002.
single nucleotide polymorphisms: a study of 2.6 million poly-
morphisms across the human genome. Genome Res 12:1679–
Zhao Z, Jiang C. 2007. Methylation-dependent transition rates are
dependent on local sequence lengths and genomic regions. Mol Biol
Zou D, Silvius D, Fritzsch B, Xu PX. 2004. Eya1 and Six1 are essential for
early steps of sensory neurogenesis in mammalian cranial placodes.
544HUMAN MUTATION 29(4),537^544,2008